U.S. patent application number 15/442270 was filed with the patent office on 2018-08-30 for system for generating calibrated multi-channel non-coherent signals.
This patent application is currently assigned to CAE INC.. The applicant listed for this patent is CAE INC.. Invention is credited to Maxime AYOTTE.
Application Number | 20180249270 15/442270 |
Document ID | / |
Family ID | 58645991 |
Filed Date | 2018-08-30 |
United States Patent
Application |
20180249270 |
Kind Code |
A1 |
AYOTTE; Maxime |
August 30, 2018 |
SYSTEM FOR GENERATING CALIBRATED MULTI-CHANNEL NON-COHERENT
SIGNALS
Abstract
A system for generating calibrated multi-channel non-coherent
signals. The system comprises a plurality of synthesizers for
generating a corresponding plurality of signals, and a plurality of
filters for band-pass filtering the plurality of generated signals.
Each filter filters the signal generated by one of the plurality of
synthesizers, by performing the band-pass filtering in a dedicated
frequency band. The system comprises a plurality of loudspeakers
for playing the plurality of filtered signals respectively filtered
by the plurality of filters. The system comprises a channel
configurator for configuring at least one of the filters for
performing the band-pass filtering according to a calibrated
amplitude spectrum of the signal. The calibrated amplitude spectrum
is determined based on a reference amplitude spectrum of the signal
and at least one of the following parameters: a target global
signal amplitude, a directionality of the signal when played by the
corresponding loudspeaker, and a frequency response of the
synthesizer.
Inventors: |
AYOTTE; Maxime;
(Saint-Laurent, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
CAE INC. |
Saint-Laurent |
|
CA |
|
|
Assignee: |
CAE INC.
Saint-Laurent
CA
|
Family ID: |
58645991 |
Appl. No.: |
15/442270 |
Filed: |
February 24, 2017 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04R 1/26 20130101; G09B
9/22 20130101; H04R 3/12 20130101; H04R 1/323 20130101; H04S
2400/11 20130101; H04R 5/02 20130101; H04S 2400/13 20130101; H04S
7/301 20130101; H04R 5/04 20130101; H04R 2499/13 20130101; H04S
7/302 20130101; H04S 3/008 20130101; G09B 9/08 20130101 |
International
Class: |
H04S 7/00 20060101
H04S007/00; H04S 3/00 20060101 H04S003/00; H04R 5/02 20060101
H04R005/02; H04R 5/04 20060101 H04R005/04; G09B 9/22 20060101
G09B009/22 |
Foreign Application Data
Date |
Code |
Application Number |
Feb 24, 2017 |
CA |
2958960 |
Claims
1. A system for generating calibrated multi-channel non-coherent
signals, comprising: a plurality of synthesizers for generating a
corresponding plurality of signals; a plurality of filters for
band-pass filtering the plurality of generated signals, each filter
filtering the signal generated by one of the plurality of
synthesizers, each filter being configured for performing the
band-pass filtering in a dedicated frequency band; a plurality of
loudspeakers for playing the plurality of filtered signals, each
loudspeaker playing the signal filtered by one of the plurality of
filters; and a channel configurator for configuring at least one of
the filters for performing the band-pass filtering according to a
calibrated amplitude spectrum of the signal, the calibrated
amplitude spectrum being determined based on a reference amplitude
spectrum of the signal and a target global signal amplitude.
2. The system of claim 1, wherein the determination of the
calibrated amplitude spectrum of the signal based on the reference
amplitude spectrum of the signal and the target global signal
amplitude comprises determining a calibrated average amplitude of
the calibrated amplitude spectrum based on a reference average
amplitude of the reference amplitude spectrum and the target global
signal amplitude, the sum of the calibrated average amplitude of
the plurality of signals being substantially equal to the target
global signal amplitude.
3. The system of claim 1, wherein the determination of the
calibrated amplitude spectrum of the signal generated by each
filter is further based on at least one of the following: a
directionality of the signal when played by the corresponding
loudspeaker and a frequency response of the synthesizer.
4. The system of claim 1, wherein each generated signal consists of
one of the following: a noise signal, a combination of noise
signals, a tone signal, a combination of tone signals, and a
combination of at least one noise signal and at least one tone
signal.
5. The system of claim 1, further comprising a physical simulation
environment of a simulator, the plurality of loudspeakers being
positioned on a wall of the physical simulation environment for
transmitting the plurality of played signals to a user of the
simulator positioned inside the physical simulation
environment.
6. The system of claim 5, wherein the simulator consists of an
aircraft simulator and at least one of the plurality of generated
signals consists of one of the following: a sound signal
corresponding to aerodynamic hiss, a sound signal corresponding to
engine noise, a sound signal corresponding to flaps being raised or
lowered, a sound signal corresponding to landing gear being
deployed or retracted, and a sound signal corresponding to a
hydraulic pump being activated.
7-18. (canceled)
Description
TECHNICAL FIELD
[0001] The present disclosure relates to the field of simulation.
More specifically, the present disclosure relates to a system for
generating calibrated multi-channel non-coherent signals in the
context of simulation.
BACKGROUND
[0002] To render simulation more realistic, noise sounds generated
during operating conditions are included in the simulation as
simulated noise sounds. Simulated noise sounds are typically
generated through filtering signals (e.g. signals captured during
operating conditions), the isolation of various sources and a
playback of the sources. Sound models outputs are mixed and sent to
a distribution mixer. For example in the context of a flight
simulator, sound models are generally further adapted with
equalization filters. However, those filters do not take into
account multiple parameters, including the simulator ambient noise,
inter-model coherent signals, channel effectiveness, channels
traveling paths, etc. Consequently, the equalization filters are
calibrated manually by the factory before shipping, to try to take
into consideration these parameters.
[0003] However, the manual factory calibration information may be
lost when multiple filters are applied on sound models. An
additional step is needed to adapt the sound models to meet a
required global sound level. If a change in the simulator ambient
noise occurs, the global sound level is affected which requires new
adjustments to be performed on the sound models. The application of
multiple filters also affects the traceability of the sound models
with the initial raw data, making updates and new adjustments
difficult to perform.
[0004] Another issue occurs when the quality of a plurality of
loudspeakers used for playing the plurality of simulated noise
sounds during simulation is not constant. In particular, the low
frequency response of the loudspeakers may vary significantly,
based on the cost and quality of each one of the loudspeakers used
for the simulation. Consequently, the rendering of sound models
having low frequencies is significantly affected by the low
frequency response of the loudspeakers in charge of playing these
sound models with low frequencies.
[0005] There is therefore a need for a new system for generating
calibrated multi-channel non-coherent signals.
SUMMARY
[0006] According to a first aspect, the present disclosure provides
a system for generating calibrated multi-channel non-coherent
signals taking into consideration a target global signal amplitude.
The system comprises a plurality of synthesizers for generating a
corresponding plurality of signals. The system comprises a
plurality of filters for band-pass filtering the plurality of
generated signals. Each filter filters the signal generated by one
of the plurality of synthesizers. Each filter is configured for
performing the band-pass filtering in a dedicated frequency band.
The system comprises a plurality of loudspeakers for playing the
plurality of filtered signals. Each loudspeaker plays the signal
filtered by one of the plurality of filters. The system comprises a
channel configurator for configuring at least one of the filters
for performing the band-pass filtering according to a calibrated
amplitude spectrum of the signal. The calibrated amplitude spectrum
is determined based on a reference amplitude spectrum of the signal
and a target global signal amplitude.
[0007] According to a second aspect, the present disclosure
provides a system for generating calibrated multi-channel
non-coherent signals, taking into consideration a directionality of
the signals when played by loudspeakers. The system comprises a
plurality of synthesizers for generating a corresponding plurality
of signals. The system comprises a plurality of filters for
band-pass filtering the plurality of generated signals. Each filter
filters the signal generated by one of the plurality of
synthesizers. Each filter is configured for performing the
band-pass filtering in a dedicated frequency band. The system
comprises a plurality of loudspeakers for playing the plurality of
filtered signals. Each loudspeaker plays the signal filtered by one
of the plurality of filters. The system comprises a channel
configurator for configuring at least one of the filters for
performing the band-pass filtering according to a calibrated
amplitude spectrum of the signal. The calibrated amplitude spectrum
is determined based on a reference amplitude spectrum of the signal
and a directionality of the signal when played by the corresponding
loudspeaker.
[0008] According to a third aspect, the present disclosure provides
a system for generating calibrated multi-channel non-coherent
signals, taking into consideration a frequency response of
synthesizers generating the signals. The system comprises a
plurality of synthesizers for generating a corresponding plurality
of signals. The system comprises a plurality of filters for
band-pass filtering the plurality of generated signals. Each filter
filters the signal generated by one of the plurality of
synthesizers. Each filter is configured for performing the
band-pass filtering in a dedicated frequency band. The system
comprises a plurality of loudspeakers for playing the plurality of
filtered signals. Each loudspeaker plays the signal filtered by one
of the plurality of filters. The system comprises a channel
configurator for configuring at least one of the filters for
performing the band-pass filtering according to a calibrated
amplitude spectrum of the signal. The calibrated amplitude spectrum
is determined based on a reference amplitude spectrum of the signal
and a frequency response of the synthesizer.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Embodiments of the disclosure will be described by way of
example only with reference to the accompanying drawings, in
which:
[0010] FIG. 1 illustrates a legacy system for generating calibrated
multi-channel coherent signals;
[0011] FIG. 2 illustrates a physical simulation environment of a
simulator using the legacy system represented in FIG. 1;
[0012] FIGS. 3A, 3B and 3C illustrate the processing by the legacy
system represented in FIG. 1 of the amplitude spectrum of model
signals;
[0013] FIG. 4 illustrates a system for generating calibrated
multi-channel non-coherent signals;
[0014] FIG. 5 illustrates the determination by the system
represented in FIG. 4 of the calibrated amplitude spectrum of the
signals, taking into consideration a target global signal
amplitude;
[0015] FIG. 6 illustrates the determination by the system
represented in FIG. 4 of the calibrated amplitude spectrum of the
signals, taking into consideration a directionality of the signals
when played loudspeakers;
[0016] FIGS. 7A and 7B illustrate the determination by the system
represented in FIG. 4 of the calibrated amplitude spectrum of the
signals, taking into consideration a frequency response of
synthesizers generating the signals;
[0017] FIGS. 8A, 8B and 8C illustrate different examples of channel
configurations;
[0018] FIG. 9 illustrates a different embodiment of the system
represented in FIG. 4 for generating different types of
signals;
[0019] FIG. 10 is schematic representation of several components of
the system represented in FIG. 4;
[0020] FIG. 11 illustrates selection of a loudspeaker based on its
low frequency rendering;
[0021] FIGS. 12A, 12B and 12C illustrate an exemplary selection by
the system of FIG. 11 of a loudspeaker among two loudspeakers
having respective low frequency responses;
[0022] FIGS. 13 and 14 further illustrate a system for dynamically
calibrating multi-channel non-coherent signals; and
[0023] FIG. 15 illustrates an algorithm implemented by a channel
configurator of the system represented in FIGS. 13 and 14.
DETAILED DESCRIPTION
[0024] The foregoing and other features will become more apparent
upon reading of the following non-restrictive description of
illustrative embodiments thereof, given by way of example only with
reference to the accompanying drawings. Like numerals represent
like features on the various drawings.
[0025] Various aspects of the present disclosure generally address
one or more of the problems related to the generation of calibrated
multi-channel non-coherent signals used in a simulator.
[0026] Throughout the present description, the following
expressions are used with relation to the corresponding
description:
[0027] Loudspeaker: a device that changes electrical signals into
sounds loud enough to be heard at a distance;
[0028] Physical simulation environment: physical space in which a
simulation is performed, for example a room, a simulator, etc.;
and
[0029] Wall: a physical delimitation either fixed or movable.
Legacy System for Generating Calibrated Multi-Channel Coherent
Signals
[0030] Referring now concurrently to FIGS. 1, 2, 3A, 3B and 3C, a
legacy system 100 for generating calibrated multi-channel coherent
signals is represented. The system 100 comprises a synthesizer 110,
a distribution mixer 120, a plurality of filters 130 and a
plurality of loudspeakers 140. The plurality of loudspeakers 140
are positioned on a wall 155 of a physical simulation environment
150 of a simulator. FIG. 2 represents a horizontal sectional view
of the physical simulation environment 150.
[0031] The entire physical simulation environment 150 and the
simulator are not represented in FIG. 2 for simplification
purposes. The term physical simulation environment shall be
interpreted in a generic manner, as a physical structure where a
user of the simulator is positioned for interacting with the
simulator during a simulation. The position of the user of the
simulator in the physical simulation environment 150 is indicated
with reference 160.
[0032] The simulator may relate to any of the following: a vehicle
simulator, a healthcare simulator, a military simulator, a mining
simulator, etc. For example, the simulator may be an aircraft
simulator, and the physical simulation environment 150 may be a
simulated aircraft cockpit.
[0033] The number of loudspeakers 140, and the position of each
loudspeaker 140 on the wall 155 of the physical simulation
environment 150, varies based on the type of simulator, based on
specific simulation needs, etc. For instance, the height of each
loudspeaker 140 on the wall 155, as well as the distance between
two neighboring loudspeakers 140, varies based on specific
simulation needs. Furthermore, a larger number of loudspeakers 140
may allow for a more realistic simulation, while a lower number of
loudspeakers 140 is more cost effective.
[0034] The position 160 of the user of the simulator is not
necessarily centralized with respect to the physical simulation
environment 150, as illustrated in FIG. 2. The user positioned at
position 160 receives a plurality of sound signals 141 respectively
played by the plurality of loudspeakers 140. The sound signals 141
simulate the real sounds that the user of the simulator would
receive when placed in real conditions corresponding to the
simulation being currently performed. For example, in the case of
the simulation of an aircraft, the sound signals 141 simulate the
real sounds generated by a corresponding real aircraft during a
takeoff phase, during a landing phase, during a high altitude
hovering phase, etc. The sound signals 141 comprise ambient noise
signals of the simulated physical environment, tone signals of the
simulated physical environment, a combination of ambient noise
signals and tone signals, etc. Examples of noise signals for an
aircraft simulator include aerodynamic hiss, the noise of engines,
etc. Examples of tone signals for an aircraft simulator include the
sound generated by flaps being raised or lowered, the sound
generated by the deployment or retracting of landing gear, the
sound generated by the activation of a hydraulic pump, etc.
[0035] The plurality of sound signals 141 played by the plurality
of loudspeakers 140 is generated by the system 100 as follows. The
synthesizer 110 generates a mixed signal composed of a plurality of
model signals mixed together. Each model signal simulates a
particular sound (e.g. a particular noise signal or a particular
tone signal) of the simulated physical environment, and has a
particular amplitude spectrum. FIG. 3A illustrates a simplified
example of amplitude spectrum of two model signals 1 and 2. FIG. 3B
illustrates a simplified example of the amplitude spectrum of the
mixed signal obtained by mixing model signal 1 and model signal 2
together.
[0036] The mixed signal is transmitted to the distribution mixer
120, which splits the mixed signal into a plurality of channels.
Each particular channel consists of a signal comprising a
particular range of frequencies of the mixed signal. For
illustrations purposes, FIG. 1 represents the splitting of the
mixed signal into three channels. However, the number of channels
may vary, and is at least equal to two. FIG. 3C illustrates a
simplified example of the splitting of the mixed signal of FIG. 3B
into three channels. Channel 1 comprises the lower frequencies
(e.g. F.sub.1 and F.sub.2 as illustrated in FIGS. 3A-C), channel 2
comprises intermediate frequencies (e.g. F.sub.3 and F.sub.4 as
illustrated in FIGS. 3A-C), and channel 3 comprises the higher
frequencies (e.g. F.sub.5 as illustrated in FIGS. 3A-C).
[0037] The signal of each channel is transmitted to an equalization
filter 130. Each equalization filter 130 is configured for
performing a band-pass filtering of the signal of the corresponding
channel in a dedicated frequency band. The filtered signal of each
channel is then transmitted to a loudspeaker 140, for playing the
filtered signal of the corresponding channel. There is one
dedicated filter 130 and one dedicated loudspeaker 140 for each
channel generated by the distribution mixer 120. The signals 141
played by the loudspeakers 140 have been previously described with
reference to FIG. 2.
[0038] The plurality of model signals mixed together by the
synthesizer 110 to generate the mixed signal are generally recorded
during real operating conditions of the physical environment to be
simulated. A library of recorded model signals is stored at the
synthesizer 110. When simulating a particular phase (e.g. takeoff,
landing, etc.), the recorded model signals corresponding to the
particular phase are mixed together for generating the mixed signal
transmitted to the distribution mixer 20.
[0039] The filters 130 are calibrated individually, so that the
combination of the sound signals 141 received by a user positioned
at position 160 has an adequate amplitude. The adequate amplitude
is determined by placing the user at position 160 and performing
the calibration of the filters 130 until the amplitude of the
combination of the sound signals 141 is adequate for performing the
simulation.
[0040] Once the calibration has been performed, an operational
phase of the simulation is performed. The operational phase
consists in having a user positioned at position 160, for
effectively performing the simulation while receiving the
combination of the sound signals 141. However, if the conditions of
the simulation vary slightly, the determined adequate amplitude may
no longer be satisfying. In this case, the calibration of the
equalization filters 130 need to be repeated, to take into
consideration the variations in the conditions of the simulation.
Since the calibration is a complex and time consuming process,
avoiding the need for manually recalibrating the equalization
filters 130 would greatly facilitate the configuration of the
simulator.
[0041] Furthermore, since the mixed signal is obtained by mixing
together a plurality of model signals, and the mixed signal is then
split into a plurality of channels, the resulting sound signals 141
played by the loudspeakers 140 in the physical simulation
environment 150 are coherent. Coherent sound signals in turn may
affect the quality of the sound simulation, by creating comb
filters (a delayed version of one of the model signals is added to
the model signal itself, causing constructive and destructive
interference).
Calibrated Multi-Channel Non-Coherent Signals
[0042] Referring now concurrently to FIGS. 2 and 4, a new system
200 for generating calibrated multi-channel non-coherent signals is
represented in FIG. 4. The system 200 comprises a plurality of
synthesizers 110, a plurality of filters 130, a plurality of
loudspeakers 140, and a channel configurator 210. As mentioned
previously with reference to FIG. 2, the plurality of loudspeakers
140 are positioned on the wall 155 of the physical simulation
environment 150 of the simulator. A first difference with the
legacy system 100 represented in FIG. 1 is that each synthesizer
110 is dedicated to a particular channel. Thus, for each particular
channel, the dedicated synthesizer 110 generates a signal which is
filtered by the corresponding filter 130, and is further played by
the corresponding loudspeaker 140. For illustrations purposes, FIG.
4 represents three synthesizers 110 for operating three
corresponding channels. However, the number of synthesizers (and
corresponding channels) may vary, and is at least equal to two. A
second difference with the legacy system 100 represented in FIG. 1
is the usage of the channel configurator 210, for automatically
configuring the signals generated by the plurality of channels. For
example, as shown on FIG. 4, the channel configurator 210
configures the filters 130 of channel 1 and channel 3, but not the
filter associated with channel 2. However, the channel configurator
210 could configure the filters 130 of all channels (i.e. channels
1, 2 and 3).
[0043] The channel configurator 210 configures the filters 130 to
perform band-pass filtering of the signal generated by the
synthesizer 110 according to a calibrated amplitude spectrum of the
signal. The calibrated amplitude spectrum is determined based on a
reference amplitude spectrum of the signal and at least one
parameter selected among one of the following: a target global
signal amplitude, a directionality of the signal when played by the
corresponding loudspeaker 140, and a frequency response of the
synthesizer 110.
[0044] The calibrated amplitude spectrum is determined by the
channel configurator 210, and the filters 130 are configured to
perform band-pass filtering of the signal generated by the
corresponding synthesizer 110 according to the calibrated amplitude
spectrum. Alternatively or concurrently, as shown on FIG. 4, the
channel configurator 210 configures the synthesizer 110 with
information allowing the synthesizer 110 to determine the
calibrated amplitude spectrum, and then generate the signal
according to the calibrated amplitude spectrum.
[0045] Referring now to FIGS. 2, 4 and 5, the determination of the
calibrated amplitude spectrum, based on the reference amplitude
spectrum of the signal and the target global signal amplitude will
be described.
[0046] For each channel N (e.g. channels 1, 2 and 3 illustrated in
FIG. 4), a model signal for simulating a particular sound (e.g. a
model noise signal or a model tone signal) of the simulator is
stored at the synthesizer 110 responsible for generating the signal
for channel N. The model signal is generally recorded during real
operating conditions, as mentioned previously in the description.
The model signal for each channel N has a reference amplitude
spectrum illustrated in FIG. 5.
[0047] The target global signal amplitude is stored by the channel
configurator 210. The target global signal amplitude is selected so
that the combination of the sound signals 141 played by the
loudspeakers 140 and received by a user positioned at position 160
in the physical simulation environment 150 has an adequate
amplitude, delay, phase and ponderation when performing the
simulation.
[0048] For each channel N, the channel configurator 210 configures
the synthesizer 110, the filter 130, or the synthesizer 110 and the
filter 130 to generate a calibrated model signal having a
calibrated amplitude spectrum determined based on the corresponding
model signal having the reference amplitude spectrum. The
configuration of the plurality of synthesizers 110 and plurality of
filters 130 takes into account the target global signal
amplitude.
[0049] Various algorithms can be implemented by the channel
configurator 210 for performing this configuration. For instance,
as illustrated in FIG. 5, for each channel N, the reference
amplitude spectrum for channel N has a reference average amplitude.
Similarly, for each channel N, the calibrated amplitude spectrum
for channel N has a calibrated average amplitude. The channel
configurator 210 determines the calibrated average amplitudes of
the channels based on the reference average amplitudes of the
channels and the target global signal amplitude, so that the sum of
the calibrated average amplitudes is substantially equal to the
target global signal amplitude. This algorithm can be applied
because the resulting signals 141 played by the loudspeakers 140
are incoherent. The determination of the calibrated average
amplitudes consists in determining a common multiplying factor, so
that when each reference average amplitude is multiplied by the
common multiplying factor for calculating the corresponding
calibrated average amplitude, the sum of the calculated calibrated
average amplitudes is substantially equal to the target global
signal amplitude. The common multiplying factor is transmitted by
the channel configurator 210 to the synthesizer 110 and/or the
filter 130, and the synthesizer 110 and/or the filter 130 uses the
common multiplying factor for generating the calibrated model
signal having the calibrated amplitude spectrum based on the
corresponding model signal having the reference amplitude
spectrum.
[0050] For each particular phase of the simulation (e.g. landing,
takeoff, etc.), only a subset of the available synthesizers 110 may
be used. Each one of the used synthesizer 110 generates one of the
calibrated model signals used for this particular phase of the
simulation. The channel configurator 210 determines which
synthesizers 110 are used; and among the used synthesizers, which
specific synthesizer 110 generates which specific calibrated model
signal with corresponding amplitude, phase and delay and how each
corresponding filter 130 is configured to perform the corresponding
band-pass filtering. A different target global signal amplitude can
be stored at the channel configurator 210 for each particular phase
of the simulation. A library of model signals covering all the
possible phases of the simulation are stored at each synthesizer
110, and calibrated model signals are generated on demand by each
synthesizer 110 under the control of the channel configurator 210.
Alternatively, the channel configurator 210 stores the library of
model signals, and transmits a particular model signal among those
present in the library to a particular synthesizer and/or to a
particular filter for generating a corresponding calibrated model
signal.
[0051] Referring now to FIGS. 2, 4 and 6, the determination of the
calibrated amplitude spectrum of the signal generated, based on the
reference amplitude spectrum of the signal and the directionality
of the signal when played by the corresponding loudspeaker 140 will
be described.
[0052] FIG. 7A illustrates an angle of incidence a determined
between the wall 155 of the physical simulation environment 150 and
the sound signal 141 played by each one of the loudspeakers
140.
[0053] The calibrated amplitude spectrum of the signal generated by
a particular synthesizer 110 and the corresponding filter 130 is
determined by modulating the reference amplitude spectrum of the
signal with the angle of incidence a determined for the
corresponding loudspeaker 140. For example, referring to FIG. 5,
the reference average amplitude of the reference amplitude spectrum
is modulated by the angle of incidence a to determine the
calibrated average amplitude of the calibrated amplitude spectrum.
An exemplary modulating factor is equal to 1/sin(.alpha.). In this
case, if the angle of incidence of the sound signal 141 is
horizontal (.alpha. is equal to 90 degrees), the modulating factor
is equal to 1. If the angle of incidence of the sound signal 141 is
not horizontal (.alpha. is greater or lower than 90 degrees), the
modulating factor is greater to 1 and increases as the angle of
incidence a increases.
[0054] Referring now to FIGS. 2, 4, 7A and 7B, the determination of
the calibrated amplitude spectrum of the signal generated by the
synthesizer 110 and filter by the corresponding filter 130, based
on the reference amplitude spectrum of the signal and a frequency
response of the synthesizer 110 will be described.
[0055] For each channel, either the synthesizer 110 and/or the
filter 130 is configured to generate the calibrated model signal.
For example, the frequency response of the synthesizer may be
static, so that a particular synthesizer 110 always applies the
same frequency response, and the corresponding filter 130 is
configured to perform band-pass filtering according to the
calibrated amplitude spectrum of the signal determined by the
channel configurator 210. Alternatively, if the frequency response
of the synthesizer 110 is dynamic, the synthesizer 110 is
configured by the channel configurator 210 with a particular
frequency response adapted to the particular model signal processed
by the synthesizer 110. The static or dynamic response is stored by
the synthesizer 110.
[0056] FIG. 7A represents the reference amplitude spectrum of the
model signal for a channel N. The synthesizer 110 in charge of
channel N is configured with the frequency response also
represented in FIG. 7A. For illustration purposes, the frequency
response is a band-pass filter only allowing the frequencies of the
model higher than F.sub.min and lower than F.sub.max. The
boundaries of the band-pass filter (F.sub.min and F.sub.max) are
configured by the channel configurator 210.
[0057] FIG. 7B represents the calibrated amplitude spectrum of the
calibrated model signal after applying the calibrated filter 130 to
the signal generated by the corresponding synthesizer 110. The
frequency F.sub.1 of the signal generated by the synthesizer 110
has been cut, while the frequencies F.sub.2 and F.sub.3 have been
allowed, by the band-pass filter 130. Thus, the calibrated
amplitude spectrum of the calibrated model signal only comprises
frequencies F.sub.2 and F.sub.3.
[0058] The band-pass filter illustrated in FIG. 7A is for
illustration purposes only. Other types of frequency responses
(e.g. low pass filter, high pass filer, etc.) can be configured,
and applied to the model signal by the filter 130 to generate the
calibrated model signal.
[0059] In a particular embodiment, the frequency response of each
filter 130 is a band-pass filter allowing frequencies in a
frequency band [F.sub.min, F.sub.max] corresponding to one third of
an octave. For instance, a first filter 130 has a frequency band
corresponding to the 1st third of a particular octave. A second
filter 130 has a frequency band corresponding to the 2nd third of
the particular octave. A third filter 130 has a frequency band
corresponding to the 3.sup.rd third of the particular octave. A
fourth filter 130 has a frequency band corresponding to the 1st
third of the octave following the particular octave, etc.
[0060] The aforementioned parameters (target global signal
amplitude, directionality of the signal when played by the
corresponding loudspeaker 140, and frequency response of the filter
130) may be combined for determining the calibrated amplitude
spectrum of the calibrated model signal generated by the
synthesizer 110, based on the reference amplitude spectrum of the
model signal. Any combination of two of the parameters, and also
the combination of the three parameters, can be used for
determining the calibrated amplitude spectrum of the calibrated
model signal. For example, as illustrated previously, the frequency
response of the synthesizer 110 is used for adjusting out
frequencies, phase and delay of the reference amplitude spectrum of
the model signal. Then, the target global signal amplitude is used
for adapting the reference average amplitude of the remaining
frequencies (the one which have not been filtered out) of the
reference amplitude spectrum of the model signal.
[0061] Referring now to FIGS. 8A, 8B, 8C and 9, several
alternatives of channel configuration are illustrated.
[0062] The signal generated by each one of the synthesizers 100
consists of one of the following: a noise signal, a combination of
noise signals, a tone signal, a combination of tone signals, and a
combination of at least one noise signal and at least one tone
signal.
[0063] FIG. 8A illustrates a synthesizer 110 generating a noise
signal, which is filtered by the corresponding filter 130 and
played by the corresponding loudspeaker 140. Alternatively, the
global noise signal played by the synthesizer 110 is a combination
of several independent noise signals.
[0064] FIG. 8B illustrates a synthesizer 110 generating a
combination of a noise signal and a tone signal, which is filtered
by the corresponding filter 130 and played by the corresponding
loudspeaker 140.
[0065] FIG. 8C illustrates a synthesizer 110 generating a noise
signal. A mixer 230 is included in the system 200 for mixing the
noise signal generated by the synthesizer 110 with a tone signal
generated by another synthesizer 220. The resulting combination of
the noise signal and the tone signal is filtered by the
corresponding filter 130 and played by the corresponding
loudspeaker 140.
[0066] Referring now to FIG. 9, a combination of synthesizer(s)
110A operating under the control of the channel configurator 210
(as illustrated in FIG. 4) and synthesizer(s) 110B operating in
combination with a distribution mixer 120 (as illustrated in FIG.
1) is represented. The synthesizer 110A is configured by the
channel configurator 210, and operates accordingly, as previously
described. The synthesizer 110A generates a calibrated model signal
on channel 1. The synthesizer 110B and the distribution mixer 120
operate as previously described. A mixed signal composed of two
model signals mixed together is generated by the synthesizer 110B,
and split by the distribution mixer 120 into two corresponding
signals transmitted on channels 2 and 3. The synthesizer 110B may
also be operating under the control of the channel configurator
210, in a manner similar to the synthesizer 110A.
[0067] The channel configurator 210 further configures the filters
130 of the channels 1 and 2 to perform band-pass filtering
according to the calibrated amplitude spectrum for received by the
channel configurator 210. As shown on FIG. 9, the present channel
configurator 210 may calibrate the filters 130 of several channels,
without necessarily calibrating the filters 130 of all
channels.
[0068] Referring now to FIG. 10, details of the channel
configurator 210, synthesizer 110 and filter 130 will be
described.
[0069] The channel configurator 210 comprises a processing unit
211, having one or more processors (not represented in FIG. 10 for
simplification purposes) capable of executing instructions of
computer program(s) (e.g. a configuration algorithm). Each
processor may further have one or several cores.
[0070] The channel configurator 210 also comprises memory 212 for
storing instructions of the computer program(s) executed by the
processing unit 211, data generated by the execution of the
computer program(s), data received via a configuration interface
213 of the channel configurator 210, etc. The channel configurator
210 may comprise several types of memories, including volatile
memory, non-volatile memory, etc.
[0071] The channel configurator 210 further comprises the
configuration interface 213. For instance, the configuration
interface 213 comprises a communication interface (e.g. a Wi-Fi
interface, an Ethernet interface, a cellular interface, a
combination thereof, etc.) for exchanging data with other entities
(such as the synthesizer 110, optionally the filter 130, a remote
computing entity, etc.) over a communication network. The
configuration interface 213 may also comprise a user interface
(e.g. a mouse, a keyboard, a trackpad, a touchscreen, etc.) for
allowing a user to interact with the channel configurator 210.
[0072] Optionally, the channel configurator 210 further comprises a
display (e.g. a regular screen or a tactile screen) for displaying
data generated by the processing unit 211.
[0073] The channel configurator 210 may be implemented by a
standard desktop or laptop computer, or by a dedicated computing
device having computing capabilities and performances.
[0074] The synthesizer 110 comprises a processing unit 111, having
one or more processors (not represented in FIG. 10 for
simplification purposes) capable of executing instructions of
computer program(s) (e.g. a configuration algorithm). Each
processor may further have one or several cores.
[0075] The synthesizer 110 also comprises memory 112 for storing
instructions of the computer program(s) executed by the processing
unit 111, data generated by the execution of the computer
program(s), data received via a configuration interface 113 of the
synthesizer 110, etc. The synthesizer 110 may comprise several
types of memories, including volatile memory, non-volatile memory,
etc.
[0076] The synthesizer 110 further comprises the configuration
interface 113. For instance, the configuration interface 113
comprises a communication interface (e.g. a Wi-Fi interface, an
Ethernet interface, a cellular interface, a combination thereof,
etc.) for exchanging data with other entities (such as the channel
configurator 210, a remote computing entity, etc.) over a
communication network.
[0077] The synthesizer 110 also comprises specialized hardware
and/or specialized software 114 for performing the generation of
the signals generated by the synthesizer 110.
[0078] The filter 130 comprises a processing unit 131, having one
or more processors (not represented in FIG. 10 for simplification
purposes) capable of executing instructions of computer program(s)
(e.g. a configuration algorithm). Each processor may further have
one or several cores.
[0079] The filter 130 also comprises memory 132 for storing
instructions of the computer program(s) executed by the processing
unit 131, data generated by the execution of the computer
program(s), data received via a configuration interface 133 of the
filter 130, etc. The filter 130 may comprise several types of
memories, including volatile memory, non-volatile memory, etc.
[0080] The filter 130 further comprises the configuration interface
133. For instance, the configuration interface 133 comprises a
communication interface (e.g. a Wi-Fi interface, an Ethernet
interface, a cellular interface, a combination thereof, etc.) for
exchanging data with other entities (such as the channel
configurator 210, a remote computing entity, etc.) over a
communication network.
[0081] The filter 130 also comprises specialized hardware and/or
specialized software 134 for performing the filtering of the
signals filtered by the filter 130 as instructed by the channel
configurator 210.
[0082] Examples of data received via the configuration interface
213 of the channel configurator 210, include: the target global
signal amplitude, the library of model signals having respective
reference amplitude spectrums, etc.
[0083] Examples of data transmitted by the configuration interface
213 of the channel configurator 210, received by the configuration
interface 113 of the synthesizer 110, and stored in the memory 112
of the synthesizer 110, include: the reference amplitude spectrum
of the model signal, the frequency response of the synthesizer 110,
the common multiplying factor for calculating the calibrated
average amplitude of the calibrated model signal based on the
reference average amplitude of the model signal, etc.
[0084] Examples of data transmitted by the configuration interface
213 of the channel configurator 210, received by the configuration
interface 133 of the filter 130, and stored in the memory 132 of
the filter 130, include: the dedicated frequency band and/or
amplitude for performing the band-pass filtering function of the
filter 130.
Selecting a Loudspeaker Based on its Low Frequency Rendering
[0085] Referring now concurrently to FIGS. 4, 11, 12A, 12B and 12C,
the system 200 for selecting a loudspeaker based on its low
frequency rendering is represented. The system 200 represented in
FIG. 11 comprises the plurality of synthesizers 110, the plurality
of filters 130, the plurality of loudspeakers 140, and the channel
configurator 210.
[0086] Each synthesizer 110 is dedicated to a particular channel.
Thus, for each particular channel, the dedicated synthesizer 110
generates a signal which is filtered by the corresponding filter
130, and is further played by the corresponding loudspeaker 140.
For illustrations purposes, FIG. 11 represents three synthesizers
110 for operating three corresponding channels. However, the number
of synthesizers (and corresponding channels) may vary, and is at
least equal to two. As mentioned previously with reference to FIG.
2, the plurality of loudspeakers 140 are positioned on the wall 155
of the physical simulation environment 150 of the simulator. The
channel configurator 210 selects a particular loudspeaker 140 based
on its low frequency rendering.
[0087] The functionalities of the channel configurator 210
represented in FIG. 4 further comprises dedicated software
component(s) (and optionally dedicated hardware component(s)) for
selecting a loudspeaker based on its low frequency rendering.
[0088] The filters 130 may be selected and/or configured so as to
have the same low frequency rendering. For example, for controlling
the cost of the system 200, a combination of lower quality and
cheaper filters 130 (with poor low frequency rendering), and higher
quality and more expensive filters 130 (with good or excellent low
frequency rendering), is used. Additionally, the system 200 may
require that some of the filers 130 have a good or excellent high
frequency rendering, a good or excellent rendering of a particular
range of frequencies, etc. Thus, each filter 130 may have a
specific frequency rendering, and in particular the low frequency
rendering of at least some of the filters 130 may differ
significantly.
[0089] If the system 200 needs to generate a model signal having a
reference amplitude spectrum with low frequencies, the channel
configurator 210 selects one among the plurality of loudspeakers
140 based on the reference amplitude spectrum of the model signal
and a low frequency response of each one of the plurality of
loudspeakers 130. The reference amplitude spectrum may only
comprise low frequencies (the model signal is a bass signal).
Alternatively, the reference amplitude spectrum comprises low
frequencies along with other frequencies. The channel configurator
210 only aims at optimizing the rendering of the low frequencies of
the reference amplitude spectrum.
[0090] FIG. 12A illustrates an exemplary low frequency response of
two loudspeakers 140. The low frequency response represents the
maximum amplitude of a signal played by each loudspeaker 140 for
frequencies included in a low frequency range. A first low
frequency response is represented for a first loudspeaker 140
associated to channel 1. A second low frequency response is
represented for a second loudspeaker 140 associated to channel 2.
Only two channels are considered for simplification purposes, but
the selection is performed based on the low frequency response of
all the channels of the system 200.
[0091] The low frequency response for the first loudspeaker 140
associated to channel 1 is good in a range of low frequencies [0,
F.sub.max]. A signal played by the first loudspeaker 140 is
rendered with a good restitution of the amplitude in the low
frequency range [0, F.sub.max].
[0092] The low frequency response for the second loudspeaker 140
associated to channel 2 is bad in the range of low frequencies [0,
F.sub.max]. A signal played by the second loudspeaker 140 is
rendered with a bad restitution of the amplitude in the low
frequency range [0, F.sub.max]. The amplitude is cut as will be
illustrated in FIG. 12B.
[0093] FIG. 12B illustrates an exemplary reference amplitude
spectrum of a model signal. The amplitude of the model signal is
represented for three exemplary frequencies (F.sub.1, F.sub.2 and
F.sub.3) included in the range of low frequencies [0, F.sub.max] of
the model signal. The amplitudes of the model signal outside the
range of low frequencies (frequencies greater than F.sub.max such
as F.sub.4 for example) is not taken into consideration by the
'.
[0094] The range of low frequencies [0, F.sub.max] taken into
consideration by the selection algorithm applied by the channel
configurator 210 may be statically configured. Alternatively, the
range of low frequencies [0, F.sub.max] taken into consideration by
the selection algorithm is dynamically adapted for each specific
model signal, based on the particular reference amplitude spectrum
of the model signal. Instead of starting at 0, the range of low
frequencies may start at a lower frequency F.sub.min and finish at
the higher frequency F.sub.max, for example 20-200 Hz.
[0095] FIG. 12C illustrates the amplitude spectrum of the model
signal played respectively by the first loudspeaker 140 (channel 1)
and the second loudspeaker 140 (channel 2).
[0096] The amplitude spectrum of the model signal played by the
first loudspeaker 140 (associated to channel 1) in the range of low
frequencies [0, F.sub.max] is the same as the reference amplitude
spectrum represented in FIG. 12B. The rendering of the model signal
in the low frequencies is adequate.
[0097] The amplitude spectrum of the model signal played by the
second loudspeaker 140 (associated to channel 2) in the range of
low frequencies [0, F.sub.max] is degraded when compared to the
reference amplitude spectrum represented in FIG. 12B (the amplitude
of the model signal at the frequencies F.sub.1, F.sub.2 and F.sub.3
has been cut by the second loudspeaker 140). The rendering of the
model signal in the low frequencies is not adequate.
[0098] In this particular example, the channel configurator 210
selects the first loudspeaker 140 associated to channel 1. Upon
selection by the channel configurator 210, the synthesizer 110
associated to channel 1 generates the model signal. Then, the
filter 130 associated to channel 1 filters the generated model
signal. Finally, the selected loudspeaker 140 (associated to
channel 1) plays the filtered model signal. The generation of the
model signal by the synthesizer 110 under the control of the
channel configurator 210 is similar to the generation of the model
signal by the synthesizer 110 represented in FIG. 4. Only two
channels have been considered for simplification purposes in this
particular example. However, the selection is performed taking into
consideration all the channels of the system 200.
[0099] An exemplary implementation of the selection algorithm
performed by the channel configurator 210 is as follows. Based on
the low frequency response of the loudspeakers 140 for all the
channels (FIG. 12A) and the reference amplitude spectrum of the
model signal (FIG. 12B), a simulated average amplitude for the low
frequencies is calculated (FIG. 12C). The simulated average
amplitude for the low frequencies consists in the average amplitude
of the reference signal when played by each loudspeaker 140
calculated over the low frequency band (e.g. [0, F.sub.max]). The
channel associated with the loudspeaker 140 providing the highest
simulated average amplitude for the low frequencies is
selected.
[0100] The memory of the channel configurator 210 stores the low
frequency response of all the loudspeakers 140, and reference
amplitude spectrums for a library of model signals which can be
generated by the synthesizers 110. Upon selection of the channel
associated with one of the loudspeakers 140 for playing a
particular model signal, the channel configurator 210 configures
via its configuration interface the synthesizer 110 associated with
the selected channel to generate the particular model signal. For
instance, the channel configurator 210 transmits via its
configuration interface the reference amplitude spectrum of the
particular model signal to the synthesizer 110 associated with the
selected channel.
Dynamically Adapting Calibrated Multi-Channel Non-Coherent
Signals
[0101] Referring now concurrently to FIGS. 13, 14 and 15, the
system 200 further dynamically adapts calibrated multi-channel
non-coherent signals.
[0102] The channel configurator 210 operates in a manner similar to
the channel configurator 210 represented in FIG. 4 for configuring
the synthesizers 110 and/or the filters 130. However, an additional
feature is added, consisting in a feedback loop from a sound sensor
420 positioned inside the physical simulation environment 150 to
the channel configurator 210. The sound sensor 420 is preferably
positioned at position 160 in the physical simulation environment
150, where a user of the simulator is positioned when performing a
simulation. Thus, the sound measured by the sound sensor 420 is as
close as possible to the sound perceived by the user performing the
simulation.
[0103] The sound measured by the sound sensor 420 is referred to as
the physical simulation environment sound. The physical simulation
environment sound comprises the plurality of sound signals 141
respectively played by the plurality of loudspeakers 140. In
addition, the physical simulation environment sound also comprises
an ambient noise created by the components of the physical
simulation environment 150. The ambient noise of the simulator is
representative of various factors, including the activity of the
user(s) of the simulator, noise created by components of the
simulator during operation, etc. The sound sensor 420 measures a
physical simulation environment signal amplitude and the ambient
noise, and transmits the physical simulation environment signal
amplitude and the measured ambient noise to the channel
configurator 210.
[0104] As mentioned previously, the channel configurator 210 of the
system 200 represented in FIG. 4 configures each synthesizer 110
and/or filter 130 to generate the signal generated by the
synthesizer 110 according to a calibrated amplitude spectrum of the
signal. The calibrated amplitude spectrum is determined based on a
reference amplitude spectrum of the signal and at least one
parameter selected among one of the following: a target global
signal amplitude, a directionality of the signal when played by the
corresponding loudspeaker 140, the ambient noise, the physical
simulation environment sound and a frequency response of the
synthesizer 110.
[0105] The channel configurator 210 of the system 200 represented
in FIG. 13 takes into consideration the feedback loop between the
sound sensor 420 and the channel configurator 210. More
specifically, the channel configurator 210 determines the
calibrated amplitude spectrum of the signal generated by the
synthesizer 110, based on the reference amplitude spectrum of the
signal and an adjusted global signal amplitude. The adjusted global
signal amplitude is calculated (by the channel configurator 210 or
the synthesizer 110) based on the target global signal amplitude,
the measured physical simulation environment signal amplitude
transmitted by the sound sensor 420 and the measured ambient noise
measured by the sound sensor 420.
[0106] The determination of the calibrated amplitude spectrum of
the signal, based on the reference amplitude spectrum of the signal
and the adjusted global signal amplitude, is similar to the
previously described determination (by the system 200) of the
calibrated amplitude spectrum of the signal, based on the reference
amplitude spectrum of the signal and the target global signal
amplitude (the target global signal amplitude is replaced by the
adjusted global signal amplitude for performing the
determination).
[0107] The target global signal amplitude is a fixed reference
value for the global signal amplitude. The target global signal
amplitude is determined for allowing the user of the simulator to
perform the simulation in the best possible audio conditions when
positioned at position 160 in the physical simulation environment
150. The target global signal amplitude is configured at the
channel configurator 210.
[0108] The physical simulation environment signal amplitude is the
signal amplitude of the sound perceived by the user of the
simulator positioned at position 160 in the physical simulation
environment 150 when performing the simulation. The physical
simulation environment signal amplitude may differ from the target
global signal amplitude based on the current operating conditions
of the system 200.
[0109] The ambient noise is the noise perceived by the user of the
simulator, and created by the immediate surroundings of the user,
such as by electric, hydraulic, pneumatic and mechanic components
of the simulator or in the vicinity of the simulator, either caused
by the operation of the simulator or independent of the operation
of the simulator.
[0110] The adjusted global signal amplitude takes into
consideration the difference between the physical simulation
environment signal amplitude, the ambient noise signal amplitude,
and the target global signal amplitude in the computation of the
calibrated amplitude spectrum of the signal generated by the
synthesizers 110 and the filters 130, so that the physical
simulation environment signal amplitude perceived by the user of
the simulator converges towards the target global signal
amplitude.
[0111] FIG. 15 illustrates an algorithm 500 for calculating the
target global signal amplitude. The steps of the algorithm 500 are
implemented by the channel configurator 210, except for step 540
which can also be implemented by the synthesizer 110.
[0112] At step 510, the stored adjusted global signal amplitude is
initialized with the target global signal amplitude. Since no
physical simulation environment signal amplitude has been taken
into consideration yet, the system 200 is supposed to be
functioning in an optimal manner.
[0113] At step 520, a new value of the physical simulation
environment signal amplitude is received from the sound sensor
420.
[0114] At step 525, a measure of the ambient noise signal in the
physical simulation environment is received.
[0115] At step 530, the adjusted global signal amplitude is
calculated by making the sum of the previously stored (at step 510
initially, and then at step 550) adjusted global signal amplitude
and the difference between the target global signal amplitude and
the received (at step 520) physical simulation environment signal
amplitude and the measured ambient noise signal.
[0116] At step 540, the calibrated amplitude spectrum of the signal
is determined based on the reference amplitude spectrum of the
signal and the calculated (at step 530) adjusted target global
signal amplitude.
[0117] At step 545, the calculated (at step 530) adjusted global
signal amplitude is stored for the next iteration.
[0118] After step 545, the next iteration of the loop starts at
step 520, as illustrated in FIG. 15.
[0119] As mentioned previously, the determination of the calibrated
amplitude spectrum of the signal based on the reference amplitude
spectrum of the signal and the adjusted global signal amplitude may
also take into consideration at least one of the following
parameters: a directionality of the signal when played by the
corresponding loudspeaker 140, and a frequency response of the
synthesizer 110.
[0120] Although the present disclosure has been described
hereinabove by way of non-restrictive, illustrative embodiments
thereof, these embodiments may be modified at will within the scope
of the appended claims without departing from the spirit and nature
of the present disclosure.
* * * * *